Victor L. Villemagne

Background The introduction of therapeutics for Alzheimer's disease has led to increased interest in precisely quantifying amyloid-β (Aβ) burden for diagnosis, treatment monitoring, and further clinical research. Recent positron emission tomography (PET) hardware innovations including digital detectors have led to superior resolution and sensitivity, improving quantitative accuracy. However, the effect of PET scanner on Centiloid remains relatively unexplored and is assumed to be minimized by harmonizing PET resolutions. Objective To quantify the differences in Centiloid between scanners in a paired cohort. Methods 36 participants from the Australian Imaging, Biomarker and Lifestyle study (AIBL) cohort were scanned within a year on two scanners. Each participant underwent 18F-NAV4694 imaging on two of the three scanners investigated, the Siemens Vision, the Siemens mCT and the Philips Gemini. We compared Aβ Centiloid quantification between scanners and assessed the effectiveness of post-reconstruction PET resolution harmonization. We further compared the scanner differences in target sub-regions and with different reference regions to assess spatial variability. Results Centiloid from the Vision camera was found to be significantly higher compared to the Gemini and mCT; the difference was greater at high-Centiloid levels. Post-reconstruction resolution harmonization only accounted for and corrected ∼20% of the Centiloid (CL) difference between scanners. We further demonstrated that residual differences have effects that vary spatially between different subregions of the Centiloid mask. Conclusions We have demonstrated that the type of PET scanner that a participant is scanned on affects Centiloid quantification, even when scanner resolution is harmonized. We conclude by highlighting the need for further investigation into harmonization techniques that consider scanner differences.

INTRODUCTION Most available p-tau217 immunoassays have similar performances. It is unclear if this is due to the use of the same antibody (the “ALZpath antibody”). We established and evaluated a novel p-tau217 assay that employs an alternative antibody, and benchmarked the results against ALZpath-p-tau217. METHODS Following development and analytical validation of the University of Pittsburgh (“Pitt-p-tau217”) method, clinical verification was performed in three independent cohorts (n=363). RESULTS Pitt-p-tau217 demonstrated high between-run stability, linearity, and specificity. Clinically, Pitt-p-tau217 differentiated neuropathologically confirmed PSEN1 mutation carriers from controls with AUC=0.94, and Aβ-PET-positive from Aβ-PET-negative cognitively normal older adults with AUC up to 0.84, equivalent to ALZpath-p-tau217 results. Both Pitt-p-tau217 and ALZpath-p-tau217 were elevated in tau-PET-positive versus tau-PET-negative participants (P=0.06; AUC=0.71 for both). Between-assay correlations were up to 0.93. DISCUSSION The new Pitt-p-tau217 assay exhibits high and reproducible classification accuracies for identifying individuals with biological evidence of AD, equivalent to the widely used ALZpath-p-tau217.

The approval of the monoclonal antibodies lecanemab (1) and donanemab (2) for the treatment of Alzheimer disease (AD) has brought a new dawn upon us. As with every dawn, it brings new hope, but it also casts long shadows. This supplement contains contributions critical to the impact of molecular imaging techniques on the implementation and assessment of newly introduced therapies, covering the detection and quantification of the underlying proteinopathy (both β-amyloid [Aβ] and tau) and brain metabolism (FDG PET), the recommended use of these new therapies, and the comparison of established fluid biomarkers (cerebrospinal fluid [CSF]) with the emerging field of plasma-based biomarkers. The issue starts with an editorial by Barthel and Drzezga (3) providing a brief appraisal of the multiple complex layers of the field of Aβ PET, its utility, and the role it plays for these novel therapies, describing the potential changes the field may undergo with the introduction of improved technologies and refined image analysis algorithms, which will lead to better quantification of the regional Aβ in the brain. This will also likely lead to a reevaluation and revision of the Centiloid standardization (4), as well as the reexamination of visual interpretation approaches. The editorial is followed by a close examination of the relationship between imaging and fluid biomarkers by the same authors (5). Above all, what we need to remember is that these 2 approaches measure 2 different biochemical pools of the protein: soluble (fluids) and insoluble (PET) (6). They are neither equivalent nor interchangeable. In the case of Aβ, both techniques reflect the same process—Aβ aggregation—but they move in opposite directions, where abnormality in CSF or plasma is marked by a decrease in Aβ, in contrast to the increase observed with Aβ PET. Further, the development of new ultrasensitive techniques, including mass spectrometry, Simoa immunoassay, and nucleic acid linked immuno-sandwich assay (known as NULISA) (7,8), allows earlier detection of abnormality with fluids than with PET (9), a technique that requires a certain density of tracer-binding sites to yield a reliable signal. Given this temporal offset, it is unfair to derive thresholds of abnormality for fluids using PET and vice versa, and novel biomarkers will need to be validated against neuropathology. On the other hand, fluids cannot make a quantitative statement on the global or regional Aβ in the brain. Also, there are differences between centrally obtained biofluids (CSF) and peripheral biofluids. CSF Aβ reflects Aβ deposition much better than plasma Aβ, (10) where only a small percentage of the Aβ in plasma actually originates in the brain. As the authors state, imaging and fluid biomarkers are complementary: fluids are much better suited for widespread screening in an efficient and economic fashion, whereas PET provides a statement of the amount and regional distribution of Aβ in the brain, allowing these biomarkers to be used, among other things, as outcome measures of drug trials or therapeutic interventions. It is also important to highlight 2 very informative figures in this paper, which illustrate beautifully what is extensively explained in the text. The first figure lists side-by-side the advantages and disadvantages of PET, CSF, and plasma biomarkers. The second figure is a flowchart of the application of the different biomarkers, showing how they should be used to determine eligibility, at least from an Aβ point of view, for therapy. The third article is an exhaustive list of recommendations of a group of experts led by Rabinovici et al. (11), detailing the appropriate use of Aβ and tau PET in clinical practice. Several clinical scenarios where Aβ or tau PET were considered and scored as “appropriate,” “uncertain,” or “rarely appropriate” are presented, with the scoring performed separately for Aβ and tau PET as stand-alone modalities. It is important to highlight and compare the appropriate use criteria in the context of anti-Aβ therapies (12,13) The article lists the approved PET tracers for Aβ and tau, outlining the visual read algorithm recommended for each tracer. As mentioned before, these tracer-specific algorithms were developed a long time ago for the purpose of obtaining regulatory approval. Now, with the dawn of anti-Aβ therapies, they will need to be reevaluated and revised, probably necessitating the development of a “universal visual read” algorithm, akin to what Centiloids (4) or CenTauR z scores (14) represent for quantification, that is more sensitive to early, asymmetric, or unusual Aβ deposition, as was already attempted by the CAPTAINs study (15). A similar approach should be adopted for tau PET, further complicated by the diverse and heterogenous presentations of tau deposition. The final article of this supplement contains specific—mostly technical—consensus guidelines for the nuclear medicine practitioner by the Society of Nuclear Medicine and Molecular Imaging and the European Association of Nuclear Medicine (16). These guidelines were designed as an educational tool, making recommendations for the acquisition, interpretation, and reporting of results of PET imaging of the brain using FDG. The most important part of the document, in the context of this supplement, relates to the application and interpretation of images in neurodegenerative conditions, ranging from AD and other dementing disorders, such as dementia with Lewy bodies or frontotemporal lobar degeneration, to movement disorders and parkinsonian syndromes (16). What is crucial in respect to FDG studies is their extreme utility in the diagnosis, prognosis, and follow-up of patients with these conditions. It is critical to emphasize the importance of FDG studies in the assessment and follow-up of neurodegenerative conditions, especially in the age of Aβ and tau PET (17). While Aβ and tau PET determine the presence and level of pathology, FDG assesses neuronal and synaptic function, so the topographic distribution of hypometabolism, especially in patients younger than 75–80 y, is an accurate and robust tool for the differential diagnosis of these neurodegenerative disorders (17). This is critical because, at the early stages of some of these dementing conditions, the clinical phenotype is similar, making it difficult to differentiate between them, which might translate into the wrong treatment. The same applies to conditions that present with identical amnestic phenotypes, like AD and limbic-predominant age-related TDP-43 encephalopathy (LATE) (18), especially in LATE cases without hippocampal sclerosis, where FDG is the only molecular imaging technique that can differentiate one from the other (19). FDG is also a robust prognostic tool. For example, glucose hypometabolism in the posterior cingulate cortex predicts cognitive decline and disease progression in patients with mild cognitive impairment (20,21). Unfortunately, FDG is not being widely used in the monitoring of anti-Aβ therapy, as it would provide a better picture of the brain function associated with Aβ removal. What we need to keep in mind is that the goal of any AD-modifying therapy is the delay of the onset of symptoms or the slowdown or arrest of disease progression, altering the disease trajectory. In other words, the purported goal is a clinical outcome, not the treatment of an image or a laboratory result, both of which are critical in determining whether the drugs are engaging the target or proving biologic—not necessarily clinical—drug efficacy. In other words, PET allows us to examine if these drugs are doing what they are supposed to do—reduce Aβ burden in the brain and prevent further Aβ accumulation—but it cannot assess a clinical outcome. The encouraging results from passive immunotherapy opens the door to new anti-Aβ therapies already in the pipeline (22), in the form of active immunization or small molecules that can be taken orally, reducing the cost of treatment while avoiding cumbersome infusions and reducing adverse effects, which will allow more widespread adoption of anti-Aβ therapies. But I mentioned shadows. These therapies have a significant, albeit modest, effect on cognitive decline (1,2). We also know now that the main driver of cognitive decline in AD is the presence of neocortical tau (23), so if a patient’s Aβ burden is very high, indicating that widespread cortical tau is already present, these anti-Aβ therapies will likely have minimal clinical effect (24). This emphasizes the importance of the information provided by tau PET, as conclusively demonstrated in the phase II and III of the donanemab trials (2,25), in which participants with high baseline cortical tau had minimal or no clinical response. This also highlights the need for earlier intervention, which will require the use of PET showing that the Aβ burden is not very high and that there is minimal or no cortical tau. Under those conditions, the recently approved therapies are likely to significantly delay the onset of symptoms or prevent the disease entirely. This is where hope lies

β-Amyloid (Αβ) imaging revolutionized the in vivo assessment of Alzheimer's disease (AD) Αβ pathology and its changes over time, increasing our insights into Aβ deposition in the brain by providing highly accurate, reliable, and reproducible quantitative statements of regional and global Aβ burden in the brain, proving essential for the differential diagnosis, staging, and evaluation of disease-specific anti-Αβ therapeutic approaches. Longitudinal observations, coupled with different disease-specific biomarkers to assess potential downstream effects of Aβ, have confirmed that Αβ deposition in the brain starts decades before the onset of symptoms. Aβ imaging studies continue to refine our understanding of the role of Αβ deposition in AD, and its relation to other imaging and fluid biomarkers. Highlights Αβ imaging revolutionized the in vivo assessment of Alzheimer's disease Αβ pathology. Αβ imaging has increased our insights into Aβ deposition in the brain by providing highly accurate, reliable, and reproducible quantitative statements of regional and global Αβ burden in the brain. Αβ imaging is essential for the differential diagnosis, staging, and evaluation of disease-specific anti-Αβ therapeutic approaches. Αβ imaging studies continue to refine our understanding of the role of Αβ deposition in Alzheimer's disease, and its relation to other imaging and fluid biomarkers.

INTRODUCTION Using the ATN framework, we evaluated the potential of plasma biomarkers to identify abnormal brain amyloid-beta (Aβ) positron emission tomography (PET), tau-PET and neurodegeneration in a socioeconomically disadvantaged population-based cohort. METHODS Community-dwelling dementia-free (n=113, including 102 (91%) cognitively normal) participants underwent ATN neuroimaging and plasma biomarker assessments. RESULTS Plasma Aβ42/Aβ40, p-tau181, and p-tau217 showed significant associations with Aβ-PET status (adjusted odds ratio [AOR] of 1.74*10-24, 1.47, and 3.43*103 respectively [p- values<0.05]), with p-tau217 demonstrating the highest classification accuracy for Aβ-PET status (AUC=0.94). Plasma p-tau181 and p-tau217 showed significant associations with tau- PET status (AOR: 1.50 and 22.24, respectively, p-values<0.05), with comparable classification accuracies for tau-PET status (AUC=0.74 and 0.70, respectively). Only plasma NfL showed significant association with neurodegeneration based on cortical thickness (AOR=1.09, p- value<0.05). CONCLUSION Our findings highlight potential of plasma p-tau217 as a biomarker for brain Aβ and tau pathophysiology, p-tau181 for tau abnormalities, and NfL for neurodegeneration in the community.